Slashdot Mirror
Intel Unveils New StrongARMs
mirko writes "Supported by the Epoc, Windows CE, RiscOS and VxWorks, the StrongARM RISC processor, which features power, low-consumption and high-frequency, could bring lots to the wireless market.
This article and this other article describe Intel's new XScale micro-architecture that will be used in the forthcoming 1GHz StrongARMs."
.. you could get some rest!
Somebody port NetBSD to these babies! Quick! =)
The right to offend is far more important than the right not to be offended. (Rowan Atkinson)
They can barely get there 1 Ghz out the door, there 1.13 Ghz seems to have lots of stability problems, and there going to try and market this? Come on, get the stuff you are already selling to work correctly and produce it in suffient volume for demand; then start annoncing other stuff. Q. Who in there right mind anounces new stuff when they can't handle their current orders. A. Marketing people for starters.
guvf vf zl fvt
sounds nice, but will it be another intel flop? -TubaMan
-TubaMan / ThE_DoOmSmItH
...PalmStation posted this article about Palm's future with the XScale technology... According to the article, "The XScale architecture offers performance roughly 20 times the Dragonball and uses less power..."
Maybe this will be foundation for PalmOS 4...
-J
Karma: T-rexcellent.
Here is another article talking about Palm's interests in these new chips. I must say that I, a PalmOS PDA owner, am very interested in the possibility of jumping from a 16MHz dragonball to a 1GHz StrongARM!
Sure it sounds nice... Doens't all new tech ?
But seriously, have anyone considered that these are RISC processors ? Do they (Intel) plan to abandon their CISC processors for the private user ? Or is this simply Intel's way of saying "we want a bigger piece of the Business pie"
I certainly think the latter is true. I seriously doubt that we could get along without the CISC, it would just cause to much incompatability, or the translation matrix would make the apparent speed increase gained from the CISC->RISC insignificant. This has no bearing on "us" the private users as I see it.
--- To err is human... Am I more human than most ?
More detail on the StrongARM range can be found on ARM's website.
Richy C.
--
Heheheh just wait until the first consumer devices roll out with BOPS DSP Core in an ARM processor. That will truely be /. worthy. Amazing the popular tech press hasn't been all over this story considering these are leaps and bounds ahead of current DSP technology, but it is not vaporware as there is working silicon!
Screw Micro$oft.
Despite the feeling of euphoria that accompanies a first post (I have felt it many times myself), remember the groundwork laid by the trolls who came before you. If you get the chance, pour out a little malt liquor to pay tribute to the fallen comrades among us.
..................................
Cunning linguists
The second article says "built using Intel's StrongARM technology". Surely they mean Advanced RISC Machines Ltd's StrongARM technology, which Intel did nothing but pay some money for.
Does my bum look big in this?
I guess the editor is probably not aware that arm processors already run a lot of deeply embedded applications. For instance, the soundchip of the dreamcast, or most cellphones. They have a 60% market share for the comms market I think. I am currently working with such a platform (lowlevel system work, and trying to fix issues in gcc as well when I can), and I must say it makes a lot of sense from a design point of view. Highest code density (especially in Thumb mode) of all 32-bit CPUs, quite efficient, and very low power consumption. No wonder it's used all over the place.
Never mind the $3 crack. This moderator has to be totally off his face on acid!
Knees (Score:1, Insightful)
sorry (Score:1, Informative)
---------------------
%46%55%43%4B !
I'm not sure who told you that RiscOS would support it immediately, but I don't think that is the case at all. The problem is that RiscOS still runs in the 24 bit mode of the very early ARM processors, while the next generation of ARM will no longer support 24 bit emulation. RiscOS Ltd have been talking about supporting 32 bit for a year or two, but as far as I know nobody is actually working on it as they don't have the resources. Perhaps if they Open Sourced their OS they would have more than a couple of developers, but that doesn't seem too likely as the RiscOS world is very deeply entrenched in the world of closed source.
what does this mean for the much-hyped Transmeta chips?
undoubtably, these StrongARM chips are Transmeta's biggest competition. Why would serious competitors in the info appliance market choose Transmeta's unproven chips instead of the market leader Intel's fast and proven chips?
Sure it sounds nice... Doens't all new tech ?
It definitely sounds nice ... I can already see the river of drool coming from people who own Risc PCs (I almost refuse to call them Acorns since Acorn sold out their workstation division but I digress).
But seriously, have anyone considered that these are RISC processors ? Do they (Intel) plan to abandon their CISC processors for the private user ? Or is this simply Intel's way of saying "we want a bigger piece of the Business pie". I certainly think the latter is true. I seriously doubt that we could get along without the CISC, it would just cause to much incompatability, or the translation matrix would make the apparent speed increase gained from the CISC->RISC insignificant. This has no bearing on "us" the private users as I see it.
I think a small history lesson is in order. The ARM architecture is not a 'new' architecture - it dates back to the mid-eighties when Acorn, having decided to skip the 16bit generation, started working on a RISC 32bit processor. In 1987, the first Acorn Archimedes was born, running Arthur OS - a fairly primitive but useable GUI and OS. This was running an ARM 2 processor at 8MHz.
Later revisions took the processor design to ARM3 with improve level 1 cache. Note these machines had no level 2 cache - as clock speed increased, this would have throttled a x86 style processor, but the ARM has fairly light memory usage as it has 13 general purpose registers, a fairly orthogonal (and small) instruction set and a load/store architecture minimising the need to go to memory for information.
Then came the ARM6 and 7 cores which took speeds up to 40MHz. At this point the ARM chips were running market leading MIPS/watt ratings - no ARM machine I have ever had has needed a heat sink - but the clock speed was starting to lag the x86 line badly. After a joint project with Digital, the StrongARM was born, screaming along at 200MHz way before the Pentiums got there, and running at less than 1W. By this time ARM Ltd had been born out of Acorn to pursue its chip dreams - but not fabrication of chips. ARM Ltd is a purely design-orientated chip creator - other partners actually build these processors. A quick trip to the ARM website will quickly show you just how widespread the ARM processor line has become - ubiquity is an almost acheived goal :-)
But just as the ARM 7's had topped out around 40MHz for a while, the 200 MHz (sometimes oc'd to 287MHz) StrongARM has remained the fastest ARM chip for a good while. During this time, DIgital got into a patent/IP dispute with Intel and ended up having to sell the StrongARM team to Intel as part of the settlement. So this is the first news of a faster ARM processor for several years - I got my 200MHz RiscPC workstation a few years ago and it blew my socks off with it's slick performance. RiscOS which is the oft preferred OS for this processor when it is used in a workstation (rather than a PDA, router, or other electronic utility) is pretty quick, and a 5x boost will be fairly insane :-) And naturally there is a port of Linux for the ARM processor (but here on Slashdot we expect nothing less).
I just regret that my RiscPC is back in the UK and I'm here in Canada with an x86. :-(
Still it would almost certainly require a motherboard upgrade ...
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
The wizzy feature that this "new microarchitecture" has seems to be no cleverer than the power/performance management which AMD already have in their chips.
It seems as if the processor (rather than PM) architecture is unchanged from the ARM core, yet they've puffed themselves up in their press release to make it sound as if the whole thing is new and improved.
Oooh, look at the price of Durons, must buy one some time...
FatPhil
Also FatPhil on SoylentNews, id 863
No need to port Linux, the article says:
The Intel XScale microarchitecture will be supported by various operating systems, including Microsoft Windows CE*, VXWorks* and IxWorks* from WindRiver, EPOC* from Symbian and Embedded Linux* from multiple vendors.
But first I should mention that there is a port of Linux: ArmLinux
The BSD one seems to be delayed.
Now, to the technical stuff.
According to Intels site, they have added power management features to the chip that allow the clock speed to be adjusted from software. Yes, this is similar to Cruesoe, but it seems like they have taken the concept even further, allowing one to go from 0 (standby) to 1000MHz. Not bad.
They have also added a few DSP functions for multimedia applications. Further details:
I dunno what all the hubub is about. These days, the strongarm is mostly used in embedded apps. For example, the Mylex ExtremeRAID cards use them and they apparently do the RAID 5 parity calculations quite quickly. If one were to quadruple the performance of the chip doing those calculations, we could end up with (finally) RAID 5 arrays that are fast enough to do meaningful work on AV projects. Cheers,
Great summary.
Whilst device penetration is huge, and ARM-powered machines from thousands of vendors, the other impressive thing is the range of licencing partners ARM have had. It's not just Digital (thence Intel), they've been in bed with other huge uP and DSP companies. For instance they had a cut-down ARM called Thumb, which was used as the master processor on things such as the Texas Instruments 7000/AV processor family - a tru multimedia chip which included several DSPs on one slab of silicon. TI could do _anything_ they want in-house - but no, they knew they'd not be able to improve on the off-the-shelf ARM core. Impressive.
FatPhil
Also FatPhil on SoylentNews, id 863
It seems that every few days or so yet another Intel story will crop up, but instead of being a real story about chips shipping or computers being made with them or actual R&D they're doing, it'll be an Announcement or Press Release about Something Really Cool (TM) that Intel will be making, shipping, and selling Real Soon Now (TM).
Please have a look and see if you can fix this. Thanks.
Mr. Ska
Although, they only ever officially pushed the StrongARM to 233MHz, and seemed to stop there for some reason.
I reckon the way Acorn could have saved themselves was by
1.Rewrite RiscOS for multiple processors (use RiscOS for home users and plums only)
2.Update their architecture to support PCI and Multiprocessors (PCI was supported in the Pheobe, but Acorn died before the Pheobe was released)
3.Make their own flavour of UNIX just for these machines (a la IRIX), and target these machines towards the SGI users. (oh, and they would have to put on decent graphics cards for this too.)
Malike Bamiyi wanted my assistance.
Their architecture was pretty good, but they insisted on keeping the OS in ROM, which i felt was not good for the developer
Er, not sure why it's not good for developers (reasons anyone?), but it's VERY good for otehr reasons - more virus and accident proof, plus it boots faster - RISCOS 4 is up and running b4 most x86 boxes are past their BIOS...
Rewrite RiscOS for multiple processors
Doubt this would have made much difference - would have just been more of a nicety I'd have thought.
Update their architecture to support PCI and Multiprocessors
Acorn were working on the first, and other people are now; multiprocessors would have been nice, but I don't think they would have made much difference.
Make their own flavour of UNIX just for these machines
Again, wouldn't have helped - linux and bsd have been ported for a while, and acorn were too small to write something like IRIX anyway. Plus, RISCOS had/has lots going for it nothing else has beaten yet.
oh, and they would have to put on decent graphics cards for this too
I'll buy this one - that would have helped!
I like how /. posters think they see through corporate media, then post press releases as "articles."
Somebody please, tell this machine I'm not a machine.
I agree than x86 assembler is a joke, I've learn many RISC assembly language and some CISC (6809, 68000, VAX), they were all easy to learn but I've never managed to surpass my disgust of the 80x86 assembly language, barf.
> Plus, 90% of the instructions are completely
> useless, so ignore them, and you got
> a reduced instruction set computer right away.
Not really, a RISC has usually lots of register, an orthogonal instruction set, a load/store architecture..
The 80x86 ISA is quite far from these kind of ISA as evidenced by its shitty FP performances due in part to its braindamaged stack-like model for FP operations.
Still there is so much competition between x86 makers that its price/performance (in the high performance area) is quite impressive (especially for integer ops).
I'm wondering: if one day Intel managed to sell only EPIC processors, it will become an Intel-only game, not more competition from AMD and the like..
Make their own flavour of UNIX just for these machines (a la IRIX), and target these machines towards the SGI users.
They had something like this around 1990. It was called RISCiX.
Note these machines had no level 2 cache - as clock speed increased, this would have throttled a x86 style processor, but the ARM has fairly light memory usage as it has 13 general purpose registers, a fairly orthogonal (and small) instruction set and a load/store architecture minimising the need to go to memory for information.
The ARM instruction set is great. It is a joy to program in ARM assembler. I especially like the possibility to add conditions to every instruction.
Nitpick: it has 16 basic registers, all of which are interchangeable. Only R15 has specific semantics (program counter).
BTW, the lack of L2 cache was (is) not a good thing. It was just never developed because at the time they couldn't get a foothold in the desktop market with the Wintel monopoly.
That's why development towards more speed was stopped in favour of extending its already amazing MIPS/Watt. They went the low-power way because the desktops weren't going anywhere -- an ARM doesn't run Windoze, after all.
You're mostly correct Toby, except for:
>During this time, DIgital got into a patent/IP
>dispute with Intel and ended up having to sell
>the StrongARM team to Intel as part of the
>settlement.
DEC did not HAVE to sell the fab that made Alpha's, StrongARM's, and PCI bridges. Although we may never know the details, the agreement between DEC and Intel was that in exchange for DEC not sueing the pants off Intel for patent infringment, Intel had to take the fab off of DEC's hands. Intel got the deal of the century. They got a first class fab, the PCI bridge chip business, AND StrongARM. Too bad for them tho that the original StrongARM team up and quit enmass.
Many within DEC thought that the CEO at the time was just prepping DEC for sale to Compaq. The rumours were that DEC had both Intel AND Microsoft dead to rights on technology that DEC developed (Alpha and VMS) and that Intel and Microsoft aquired without license.
As I said, we'll probably never know the details. And that's a shame.
ok i must admit there are several times i've completely fucked a riscpc, the been able to just do a ctrl-x-break-del and have it all fine again
Malike Bamiyi wanted my assistance.
The ARM instruction set is great. It is a joy to program in ARM assembler. I especially like the possibility to add conditions to every instruction.
Yes - conditional execution is a major plus and probably helps the ARM assembler get such good code density.
Nitpick: it has 16 basic registers, all of which are interchangeable. Only R15 has specific semantics (program counter).
Major Nitpick: Two other registers are vulnerable - R14 is used as a link register for branches and I counted it out. It's also dangerous to fiddle with this when switching IRQ, FIQ or SVC modes as it changes - the same is true for R13. That is why I gave the 13 general purpose registers r0-r12 - I never used r13-r15 except in special cases :-) Really the ARM has more general purpose registers than 16 - there are 27 (ARM 2 and 3) or 31 (ARM6+) registers but you only see 16 at any one time - changes in processor mode can swap registers over.
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
That 0.60 is variable by task. Some tasks its closer to 1.0, some its closer to 0.10 (we just don't do those
So, lets look at the Intel XScale Benchmarks or rather hallucinatory benchmarks since they don't have the silicon yet...
It looks like a 1GHz Xscale is about three times as fast as a 233MHz StrongARM. Thats three times as fast as a Pentium III 130MHz, or perhaps in the ballpark of a PentiumIII 400MHz.
So stop having GHz envy and instead marvel at the really neat parts of the architecture.
(Incidentally, don't b*tch at me about my 0.60 estimate. email me and I'll give you an account on a benchmark machine and you can't run your own. If you can figure out how to email me then you aren't worthy.)
P.S. It's RISC OS. There is another OS called RiscOS which has nothing to do with Acorn/Pace/whatever
Does anyone have a pointer to the new ISA ?
:-(
:-)
All the articles linked until now are quite vague and non-technical..
Will they be as elegant as the Altivec ISA ?
Yes, I do think that some ISA are more elegant than others ARM ISA is quite nice and I like conditional operations
Bill - aka taniwha
--
Bill - aka taniwha
--
Leave others their otherness. -- Aratak
they have all already been ported, they run on an Acorn RISC PC or a NetWinder (AFAIK), both StrongARM powered
Intel got a .35 micron fab when the industry standard was .25, and Intel was moving to .22. That's not what I call first class. Intel had already made PCI chipsets for a while; acquiring DEC didn't vastly change that. And of course the StrongARM team quit enmass -- Intel is a 'hire-em-young-&-screw-em-while-theyre-naive' company, just lke microsoft (and you wonder why both companies have poor quality control). Last I heard a big chunk of the StrongARM guys went to Candence, or some other tools company. Bummer.
Bill - aka taniwha
--
Bill - aka taniwha
--
Leave others their otherness. -- Aratak
Where the f***k all my previous posts went? It seems that moderator/cencor/whoeveryouare do not like my comments and wants to keep his job badly, especially after the doodoo with the Kerberos protocol. Heh, screw you.
I'd really like to buy a StongARM system if only I could buy the CPU/MOBO in a standard AT/ATX form factor. Anyone know of a modern ARM based system (short of the now dead Netwinder) which can be had on the cheap and uses commodity parts?
morgus morphus opined:
I don't know about FreeBSD (that page was last updated in 1995!). However, there is a NetBSD ARM page here.
According to that page:
They have a short history of the NetBSD/arm32 project.
---
In a hundred-mile march,
The DSP unit is basically a 2 way SIMD integer MAC unit. Nothing as sofisticated (some might say bloated...) as Altivec.
One poster wrote that the StrongARM is only about .60 the speed of a PentiumIII at the same clock. That number varied and would even go down to .10.
But at 1.6W and only costing around $20, how many of these could you put together to summarily spank the Pentium performance-wise and still run cooler and leave more in my pocket? Are there any workstations that already have multiple StrongARMs?
Aah, change is good. -- Rafiki
Yeah, but it ain't easy. -- Simba
Right here.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
P.S.
Netwinder is FAR from dead.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
With all of this interesting talk of ARM performace, I am curious to know if this ARM architecture is capable of being used for the purpose of Symmetric Multiprocessing . Is there anybody that can offer insight to this question? It is neat to think that you could have an 8-processor system, that performs nicely, and gives off virtually no heat.
As mentioned in a previous Slashdot article the Compaq iPaq uses a StrongARM processor and runs Linux. I wonder if we can build a small daughter card that drops the Xscale (tiny, BGA package) into the space from a desoldered SA110 (inch square, PQFP package) with the necessary bus conversion so we don't have to wait for Compaq to give us a 1BIPS Linux palm-top with 36 hours battery life... I better go out and buy a really small tip for my soldering iron :-)
If intelligent life is too complex to evolve on its own, who designed God?
The Interl site doesn't mention anything about a cache-coherent bus protocol, so I assume that these new CPUs don't support SMP.
Many within DEC thought that the CEO at the time was just prepping DEC for sale to Compaq. The rumours were that DEC had both Intel AND Microsoft dead to rights on technology that DEC developed (Alpha and VMS) and that Intel and Microsoft aquired without license.
Supposedly the Intel Pentium was using branch predict algorithms that were in the Alpha when DEC was trying to see who they can get to support this technology, and apparently Intel stole some details that they disclosed.
For the Microsoft case, I think the lead designer for VMS went to Microsoft and made a very VMS-like OS there, then called Windows NT. At the surface, the similarities were few (some apps were named the same and performed the same function), but the underlying architecture was said to be very similar. There was talk that that lead designer took a lot of trade secrets and other intellectual property with him and put it in NT.
Since then, Dec / Compaq has licenced out the technologies with in VMS to Microsoft to integrate into what was then slated to be called NT 6. At this rate, that product will probably Windows 2010.
It was easy to find out how to email you.
Google.
Your name seems reasonably unique, and therefore I can say with confidence that the address is
jim
at
federated
dot
com
Google is the answer.
Most of the StrongARM processors used in portable devices rarely exceed 200MHz. And now talk of a 1 GHz StrongARM? BRING IT ON!
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer
One small correction to the timeline: Apple Computer used the ARM CPUs in the Newton PDA (which Steve Jobs killed off after he took over again - not one of his better decisions).
It was Apple that insisted that Acorn Computer divest itself of the ARM development team, so that they could be buying from a supplier that wasn't directly competing with them in the computer systems market. Thus was Acorn RISC Machines (ARM), Ltd. born.
The collaboration with DEC came later, and that produced the StrongARM.
By the way, my Win2k machine is still up and running at full capacity. How's your computer, still breathing the bilgewater of the internet?
"Ancillary does not mean you get to rule the world." --U.S. Circuit Judge Harry Edwards, speaking to the FCC's lawyer